DESCRIPTION (Applicant's Description): Cells communicate with each other during development to create positional information that is used to establish different cell fates. A relatively small number of distinct signaling systems converge within the cell to modify the activity of pre-existing transcription factors, which bind in a combinatorial fashion to cis-acting enhancer elements and regulate the expression of downstream target genes. In contrast to converging signal transduction pathways and integrating enhancer elements that function within the receiving cell, relatively little is known about how signaling is controlled extracellularly prior to receptor activation. The focus of this proposal is to understand how various forms of the Drosophila Sog protein function extracellularly to regulate BMP signaling during development of the embryonic neural and non-neural ectoderm and adult wing pattern. We present preliminary data indicating that the short gastrulation (sog) gene, which encodes a large extracellular protein (Sog), can modulate the activities of distinct members of the Bone Morphogenetic Protein (BMP) subfamily of TGF-beta ligands such as Decapentaplegic (Dpp) and Glass bottom boat (Gbb). Our analysis suggests that the full length Sog protein is processed in a cell-type specific fashion to generate different forms of Sog which block signaling mediated by distinct subsets of BMPs. For example, a truncated form of Sog (Supersog) has broader BMP inhibitory activity than intact Sog. Supersog-like molecules are made in vivo, and this processing can be reconstituted in vitro by combining intact Sog with the metalloprotease Tolloid (Tld) and a co-factor known as Twisted Gastrulation (Tsg). The overall goal of the experiments proposed in this application is to understand the developmental function of different Sog activities, analyze the mechanism by which alternative forms of Sog are produced, and determine how these molecules modulate signaling by different BMP ligands at short or long range. The Specific Aims of this proposal are: 1) Test the model that Tld plus Tsg processes Sog into Supersog. 2) Determine whether different forms of Sog block selected subsets of BMP ligands. 3) Analyze positive acting forms of Sog and determine their range of action.